4 papers
Preference Packing: Efficient Preference Optimization for Large Language Models
Jaekyung Cho
Resource-efficient training optimization techniques are becoming increasingly important as the size of large language models (LLMs) continues to grow. In particular, batch packing…
Beyond RAG vs. Long-Context: Learning Distraction-Aware Retrieval for Efficient Knowledge Grounding
Seongwoong Shim, Myunsoo Kim, Jae Hyeon Cho +1
Retrieval-Augmented Generation (RAG) is a framework for grounding Large Language Models (LLMs) in external, up-to-date information. However, recent advancements in context window s…
K/DA: Automated Data Generation Pipeline for Detoxifying Implicitly Offensive Language in Korean
Minkyeong Jeon, Hyemin Jeong, Yerang Kim +3
Language detoxification involves removing toxicity from offensive language. While a neutral-toxic paired dataset provides a straightforward approach for training detoxification mod…
VPO: Leveraging the Number of Votes in Preference Optimization
Jae Hyeon Cho, Minkyung Park, Byung-Jun Lee
Direct Preference Optimization (DPO) trains a language model using human preference data, bypassing the explicit reward modeling phase of Reinforcement Learning from Human Feedback…